Model za ocenjevanje vrednosti projektov razvoja spletnih programskih rešitev

Milinković, Nina

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Model za ocenjevanje vrednosti projektov razvoja spletnih programskih rešitev
Authors:	ID Milinković, Nina (Author) ID Kern, Tomaž (Mentor) More about this mentor...
Files:	MAG_Milinkovic_Nina_2024.pdf (3,95 MB) MD5: D0C1377A66AE47F8E719429C8EEFC17E
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FOV - Faculty of Organizational Sciences in Kranj
Abstract:	V nalogi predstavljamo zasnovo in implementacijo modela multiple linearne regresije v »Podjetju X«, ki se ukvarja z razvojem spletnih programskih rešitev. Model je bil zasnovan za izboljšanje natančnosti pri napovedovanju presežka ur v projektih, kar omogoča boljše planiranje in oceno projektov. S sistematično analizo podatkov iz preteklih projektov in oblikovanjem vprašalnika, ki upošteva ključna področja, kot so «inovativne rešitve in razvoj«, «upravljanje in vodenje«, »infrastruktura in procesi« ter »ocenjevanje in odnosi s strankami«, in pripadajoče parametre smo razvili model, ki omogoča natančnejše planiranje in ocenjevanje prihodnjih projektov. Model multiple linearne regresije smo razvili v skladu s smernicami za razvoj linearnih regresijskih modelov, ki vključujejo preverjanje sedmih ključnih predpostavk ustreznosti modela, ki smo jih v nalogi potrdili. Za validacijo modela smo uporabili metodo validacije s pomočjo prilagojenega koeficienta determinacije in validacijo ponovnega vzorčenja z zamenjavo –, model pa smo testirali na dejanskem primeru projekta v »Podjetju X«. Rezultati validacije s statističnimi metodami kažejo na statistično značilne korelacije v obeh vzorcih, kar potrjuje, da so napovedi modela tesno usklajene z dejanskimi podatki. Validacija na konkretnem primeru pa je pokazala, da uvedba modela multiple linearne regresije uspešno napoveduje presežek ur že pred začetkom projekta. Z napovedjo, ki od dejanskega stanja glede presežka ur odstopa za le 3 odstotke, model zmanjšuje izgube zaradi neoptimalnega napovedovanja ur pri projektih in podjetju potencialno omogoča ustvarjanje dodatnega zaslužka – če seveda podjetje dodatne ure proda.
Keywords:	Planiranje in ocenjevanje vrednosti projekta, model multiple linearne regresije, projektni vodja, spletne programske rešitve.
Place of publishing:	Maribor
Year of publishing:	2024
PID:	20.500.12556/DKUM-91067
COBISS.SI-ID:	216096003
Publication date in DKUM:	05.06.2025
Views:	127
Downloads:	20
Metadata:
Categories:	FOV
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	18.10.2024

Secondary language

Language:	English
Title:	Model for evaluating the value of web development solutions projects
Abstract:	Master's thesis presents the design and implementation of a multiple linear regression model at »Company X«, which specializes in the development of web software solutions. The model was designed to improve accuracy in predicting project hour overruns, enabling better project planning and assessment. Through systematic analysis of data from past projects and the creation of a questionnaire that considers key areas such as »innovative solutions and development«, »management and leadership«, »infrastructure and processes«, and »evaluation and customer relations«, along with their associated parameters, we developed a model that allows for more accurate planning and evaluation of future projects. The multiple linear regression model was developed in accordance with guidelines for developing linear regression models, which include verifying seven key assumptions of model validity, all of which were confirmed in this paper. To validate the model, we used methods involving the adjusted coefficient of determination and resampling validation with replacement. The model was tested on an actual project case in »Company X«. The validation results, based on statistical methods, show statistically significant correlations in both samples, confirming that the model's predictions are closely aligned with the actual data. Validation on a real-world case demonstrated that the introduction of the multiple linear regression model successfully predicts hour overruns even before the project begins. With predictions deviating from the actual overrun by only 3 percent, the model reduces losses due to suboptimal hour predictions in projects and potentially enables the company to generate additional revenue—assuming, of course, the company sells the additional hours.
Keywords:	Project planning and evaluation, multiple linear regression model, project manager, web software solutions.

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back